ollama: thinking, tool streaming, docs, tests (#31772)

* New `reasoning` (bool) param to support toggling [Ollama
thinking](https://ollama.com/blog/thinking) (#31573, #31700). If
`reasoning=True`, Ollama's `thinking` content will be placed in the
model responses' `additional_kwargs.reasoning_content`.
  * Supported by:
    * ChatOllama (class level, invocation level TODO)
    * OllamaLLM (TODO)
* Added tests to ensure streaming tool calls is successful (#29129)
* Refactored tests that relied on `extract_reasoning()`
* Myriad docs additions and consistency/typo fixes
* Improved type safety in some spots

Closes #29129
Addresses #31573 and #31700
Supersedes #31701

libs/partners/ollama/tests/unit_tests/test_chat_models.py

@@ -23,7 +23,7 @@ class TestChatOllama(ChatModelUnitTests):
 
     @property
     def chat_model_params(self) -> dict:
-        return {"model": "llama3-groq-tool-use"}
+        return {"model": MODEL_NAME}
 
 
 def test__parse_arguments_from_tool_call() -> None:
@@ -51,7 +51,6 @@ def test_arbitrary_roles_accepted_in_chatmessages(
     monkeypatch.setattr(Client, "stream", _mock_httpx_client_stream)
 
     llm = ChatOllama(
-        base_url="http://whocares:11434",
         model=MODEL_NAME,
         verbose=True,
         format=None,

libs/partners/ollama/tests/unit_tests/test_embeddings.py

@@ -10,7 +10,7 @@ MODEL_NAME = "llama3.1"
 
 def test_initialization() -> None:
     """Test embedding model initialization."""
-    OllamaEmbeddings(model="llama3", keep_alive=1)
+    OllamaEmbeddings(model=MODEL_NAME, keep_alive=1)
 
 
 @patch("langchain_ollama.embeddings.validate_model")

libs/partners/ollama/tests/unit_tests/test_llms.py

@@ -10,25 +10,25 @@ MODEL_NAME = "llama3.1"
 
 def test_initialization() -> None:
     """Test integration initialization."""
-    OllamaLLM(model="llama3")
+    OllamaLLM(model=MODEL_NAME)
 
 
 def test_model_params() -> None:
     # Test standard tracing params
-    llm = OllamaLLM(model="llama3")
+    llm = OllamaLLM(model=MODEL_NAME)
     ls_params = llm._get_ls_params()
     assert ls_params == {
         "ls_provider": "ollama",
         "ls_model_type": "llm",
-        "ls_model_name": "llama3",
+        "ls_model_name": MODEL_NAME,
     }
 
-    llm = OllamaLLM(model="llama3", num_predict=3)
+    llm = OllamaLLM(model=MODEL_NAME, num_predict=3)
     ls_params = llm._get_ls_params()
     assert ls_params == {
         "ls_provider": "ollama",
         "ls_model_type": "llm",
-        "ls_model_name": "llama3",
+        "ls_model_name": MODEL_NAME,
         "ls_max_tokens": 3,
     }